Path: csiph.com!usenet.pasdenom.info!weretis.net!feeder1.news.weretis.net!feeder.erje.net!news.internetdienste.de!news.tu-darmstadt.de!news.belwue.de!newsfeed.ision.net!newsfeed2.easynews.net!ision!newsfeed.arcor.de!newsspool1.arcor-online.net!news.arcor.de.POSTED!not-for-mail
Content-Type: text/plain; charset="UTF-8"
Message-ID: <1703829.TrCZmHXVCy@PointedEars.de>
From: Thomas 'PointedEars' Lahn <PointedEars@web.de>
Reply-To: Thomas 'PointedEars' Lahn <cljs@PointedEars.de>
Organization: PointedEars Software (PES)
Date: Wed, 14 Mar 2012 00:17:53 +0100
User-Agent: KNode/4.4.11
Content-Transfer-Encoding: 8Bit
Subject: Re: on*= Clauses: When to return
Newsgroups: comp.lang.javascript
References: <a84al79rveq3549usk0q3et55uubgkmbu1@4ax.com> <86ef4e24-2765-4403-884f-18cce40028f1@d17g2000vba.googlegroups.com> <htdtl7tir8gr9e877ifjsmnhfdd7fqnsj7@4ax.com> <c39119be-4fef-4310-93e9-5eee6ea66a06@b18g2000vbz.googlegroups.com> <4sltl7ldfha1ojnssmiv6l08ci1d34rtic@4ax.com> <c318b617-dbda-424b-9cd4-ec6046301ba0@r34g2000yqn.googlegroups.com> <farul71evdljlqqueevha0isd2ivtac457@4ax.com> <9857e04b-250d-4dff-990c-8ee893b8ae7a@ow8g2000pbc.googlegroups.com> <qq8vl7hjg9j2f4kfl6tt0mbcpbt1m7dcqn@4ax.com> <8b2b138b-7f2d-4705-ae5c-701fdf60bc1e@q12g2000yqg.googlegroups.com>
Followup-To: comp.lang.javascript
MIME-Version: 1.0
Lines: 78
NNTP-Posting-Date: 14 Mar 2012 00:17:54 CET
NNTP-Posting-Host: ecc4878a.newsspool2.arcor-online.net
X-Trace: DXC=^JGV^h<fn]RYI9]OHn9o5^A9EHlD;3YcR4Fo<]lROoRQ8kF<OcfhCO[8`\NS>d\^5XDZm8W4\YJN\b@mF9jNikdU@_VbHgL8kiQXobhA7RNfDZ
X-Complaints-To: usenet-abuse@arcor.de
Xref: csiph.com comp.lang.javascript:12664

David Mark wrote:

> Gene Wirchenko wrote:
>> David Mark wrote:
>> >On Mar 13, 11:56 am, Gene Wirchenko <ge...@ocis.net> wrote:
>>
>> [snip]
>>
>> >You can't prevent the blur, but you can try to simulate that. Good
>> >luck.
>>
>> No luck.  Already done.  I am coding for use in IE 9 only for
>> now, and it works reliably.
> 
> Okay.
> 
>> Well, I may need a bit of luck later.  I recently found out that
>> my focus reset does not work in Firefox 10.
> 
> I imagine you were braced for that. :)

A correct assumption.

>> I looked at one possible
>> culprit, but that was not it.  Something for a bit later.
> 
> It's difficult if you don't understand exactly why something works (or
> doesn't). Too often in this industry, developers consider anything
> that appears to work in a subset of "all browsers" to be correct. In
> other words, conflating logic with observation. It leads to nothing
> but aggravation when something "breaks" and you don't even know why it
> was "working" in the first place. ;)
> […]
>> >> It is required for decent data
>> >> entry according to the way that we like it.
>>
>> >> I find silly the idea that DE methods that have worked for years
>> >> and years have to be tossed <fanfare> because of the Web.
>>
>>> AFAIK, nobody ever thought that was a good idea in GUI design.
>> Which one are you talking about?
> 
> Stealing the focus, particularly after a tab.

JFTR: This is a typical beginner's mistake, and so is the notion that it 
would be appropriate if web code only worked in one (version of) one 
browser.

"Intranet", you say?  Intranets do change, and the more they depend on 
features or behavior of a particular browser or browser version, the more 
expensive is the adoption of another, probably faster and more secure 
browser, as the planned small changes will gradually evolve into a full 
rewrite.  I have been there before.

As for calling focus() on blur, precisely because that is bad UI design, 
working against the user's expectations and making a web form inaccessible, 
the Gecko code contains precautions to prevent trapping the user in a 
control.  It's not a bug, it's a feature.  And if IE does not have it by 
now, it is going to; perhaps in version 10.

The more wannabes abuse features, and the more users are complaining about 
that, the more likely is it that vendors will make efforts to prevent that 
kind of abuse.  This can be seen, for example, when comparing the user 
preferences dialogs to restrict certain aspects of web scripting between 
Firefox 1 and Firefox 10: the latter does not even have the former's option 
to prevent scripts from changing the status bar text anymore, as that is 
disallowed by default now.  Likewise for other browsers and versions.

Who writes code on the basis that it is working reliably today, is writing 
code for the trash.


PointedEars
-- 
    realism:    HTML 4.01 Strict
    evangelism: XHTML 1.0 Strict
    madness:    XHTML 1.1 as application/xhtml+xml
                                                    -- Bjoern Hoehrmann